lib/scripts/layout2layout.py

   1 #! /usr/bin/env python
   2 # -*- coding: utf-8 -*-
   3
   4 # file layout2layout.py
   5 # This file is part of LyX, the document processor.
   6 # Licence details can be found in the file COPYING.
   7
   8 # author Georg Baum
   9
  10 # Full author contact details are available in file CREDITS
  11
  12 # This script will update a .layout file to current format
  13
  14
  15 import os, re, string, sys
  16
  17 # Incremented to format 4, 6 April 2007, lasgouttes
  18 # Introduction of generic "Provides" declaration
  19
  20 # Incremented to format 5, 22 August 2007 by vermeer
  21 # InsetLayout material
  22
  23 # Incremented to format 6, 7 January 2008 by spitz
  24 # Requires tag added to layout files
  25
  26 # Incremented to format 7, 24 March 2008 by rgh
  27 # AddToPreamble tag added to layout files
  28
  29 # Incremented to format 8, 25 July 2008 by rgh
  30 # UseModule tag added to layout files
  31 # CopyStyle added to InsetLayout
  32
  33 # Incremented to format 9, 5 October 2008 by rgh
  34 # ForcePlain and CustomPars tags added to InsetLayout
  35
  36 # Incremented to format 10, 6 October 2008 by rgh
  37 # Change format of counters
  38
  39 # Incremented to format 11, 14 October 2008 by rgh
  40 # Add ProvidesModule, ExcludesModule tags
  41
  42 # Incremented to format 12, 10 January 2009 by gb
  43 # Add I18NPreamble tag
  44
  45 currentFormat = 12
  46
  47
  48 def usage(prog_name):
  49     return ("Usage: %s inputfile outputfile\n" % prog_name +
  50             "or     %s <inputfile >outputfile" % prog_name)
  51
  52
  53 def error(message):
  54     sys.stderr.write(message + '\n')
  55     sys.exit(1)
  56
  57
  58 def trim_eol(line):
  59     " Remove end of line char(s)."
  60     if line[-2:-1] == '\r':
  61         return line[:-2]
  62     elif line[-1:] == '\r' or line[-1:] == '\n':
  63         return line[:-1]
  64     else:
  65         # file with no EOL in last line
  66         return line
  67
  68
  69 def read(input):
  70     " Read input file and strip lineendings."
  71     lines = list()
  72     while 1:
  73         line = input.readline()
  74         if not line:
  75             break
  76         lines.append(trim_eol(line))
  77     return lines
  78
  79
  80 def write(output, lines):
  81     " Write output file with native lineendings."
  82     for line in lines:
  83         output.write(line + os.linesep)
  84
  85
  86 # Concatenates old and new in an intelligent way:
  87 # If old is wrapped in ", they are stripped. The result is wrapped in ".
  88 def concatenate_label(old, new):
  89     # Don't use strip as long as we support python 1.5.2
  90     if old[0] == '"':
  91         return old[0:-1] + new + '"'
  92     else:
  93         return '"' + old + new + '"'
  94
  95 # appends a string to a list unless it's already there
  96 def addstring(s, l):
  97     if l.count(s) > 0:
  98         return
  99     l.append(s)
 100
 101
 102 def convert(lines):
 103     " Convert to new format."
 104     re_Comment = re.compile(r'^(\s*)#')
 105     re_Counter = re.compile(r'\s*Counter\s*', re.IGNORECASE)
 106     re_Name = re.compile(r'\s*Name\s+(\S+)\s*', re.IGNORECASE)
 107     re_UseMod = re.compile(r'^\s*UseModule\s+(.*)', re.IGNORECASE)
 108     re_Empty = re.compile(r'^(\s*)$')
 109     re_Format = re.compile(r'^(\s*)(Format)(\s+)(\S+)', re.IGNORECASE)
 110     re_Preamble = re.compile(r'^(\s*)Preamble', re.IGNORECASE)
 111     re_EndPreamble = re.compile(r'^(\s*)EndPreamble', re.IGNORECASE)
 112     re_MaxCounter = re.compile(r'^(\s*)(MaxCounter)(\s+)(\S+)', re.IGNORECASE)
 113     re_LabelType = re.compile(r'^(\s*)(LabelType)(\s+)(\S+)', re.IGNORECASE)
 114     re_LabelString = re.compile(r'^(\s*)(LabelString)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 115     re_LabelStringAppendix = re.compile(r'^(\s*)(LabelStringAppendix)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 116     re_LatexType = re.compile(r'^(\s*)(LatexType)(\s+)(\S+)', re.IGNORECASE)
 117     re_Style = re.compile(r'^(\s*)(Style)(\s+)(\S+)', re.IGNORECASE)
 118     re_CopyStyle = re.compile(r'^(\s*)(CopyStyle)(\s+)(\S+)', re.IGNORECASE)
 119     re_NoStyle = re.compile(r'^(\s*)(NoStyle)(\s+)(\S+)', re.IGNORECASE)
 120     re_End = re.compile(r'^(\s*)(End)(\s*)$', re.IGNORECASE)
 121     re_Provides = re.compile(r'^(\s*)Provides(\S+)(\s+)(\S+)', re.IGNORECASE)
 122     re_CharStyle = re.compile(r'^(\s*)CharStyle(\s+)(\S+)$', re.IGNORECASE)
 123     re_AMSMaths = re.compile(r'^\s*Input amsmaths.inc\s*')
 124     re_AMSMathsPlain = re.compile(r'^\s*Input amsmaths-plain.inc\s*')
 125     re_AMSMathsSeq = re.compile(r'^\s*Input amsmaths-seq.inc\s*')
 126     re_TocLevel = re.compile(r'^(\s*)(TocLevel)(\s+)(\S+)', re.IGNORECASE)
 127
 128     # counters for sectioning styles (hardcoded in 1.3)
 129     counters = {"part"          : "\\Roman{part}",
 130                 "chapter"       : "\\arabic{chapter}",
 131                 "section"       : "\\arabic{section}",
 132                 "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 133                 "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 134                 "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 135                 "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 136
 137     # counters for sectioning styles in appendix (hardcoded in 1.3)
 138     appendixcounters = {"chapter"       : "\\Alph{chapter}",
 139                         "section"       : "\\Alph{section}",
 140                         "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 141                         "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 142                         "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 143                         "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 144
 145     # Value of TocLevel for sectioning styles
 146     toclevels = {"part"          : 0,
 147                  "chapter"       : 0,
 148                  "section"       : 1,
 149                  "subsection"    : 2,
 150                  "subsubsection" : 3,
 151                  "paragraph"     : 4,
 152                  "subparagraph"  : 5}
 153
 154     i = 0
 155     only_comment = 1
 156     counter = ""
 157     toclevel = ""
 158     label = ""
 159     labelstring = ""
 160     labelstringappendix = ""
 161     space1 = ""
 162     labelstring_line = -1
 163     labelstringappendix_line = -1
 164     labeltype_line = -1
 165     latextype = ""
 166     latextype_line = -1
 167     style = ""
 168     maxcounter = 0
 169     format = 1
 170     formatline = 0
 171     usemodules = []
 172
 173     while i < len(lines):
 174         # Skip comments and empty lines
 175         if re_Comment.match(lines[i]) or re_Empty.match(lines[i]):
 176             i += 1
 177             continue
 178
 179         # insert file format if not already there
 180         if (only_comment):
 181             match = re_Format.match(lines[i])
 182             if match:
 183                 formatline = i
 184                 format = int(match.group(4))
 185                 if format > 1 and format < currentFormat:
 186                     lines[i] = "Format %d" % (format + 1)
 187                     only_comment = 0
 188                 elif format == currentFormat:
 189                     # nothing to do
 190                     return format
 191                 else:
 192                     error('Cannot convert file format %s' % format)
 193             else:
 194                 lines.insert(i, "Format 2")
 195                 only_comment = 0
 196                 continue
 197
 198         # Don't get confused by LaTeX code
 199         if re_Preamble.match(lines[i]):
 200             i += 1
 201             while i < len(lines) and not re_EndPreamble.match(lines[i]):
 202                 i += 1
 203             continue
 204
 205         # This just involved new features, not any changes to old ones
 206         if format == 11:
 207           i += 1
 208           continue
 209
 210         if format == 10:
 211             match = re_UseMod.match(lines[i])
 212             if match:
 213                 module = match.group(1)
 214                 lines[i] = "DefaultModule " + module
 215             i += 1
 216             continue
 217
 218         if format == 9:
 219             match = re_Counter.match(lines[i])
 220             if match:
 221                 counterline = i
 222                 i += 1
 223                 while i < len(lines):
 224                     namem = re_Name.match(lines[i])
 225                     if namem:
 226                         name = namem.group(1)
 227                         lines.pop(i)
 228                         lines[counterline] = "Counter %s" % name
 229                         # we don't need to increment i
 230                         continue
 231                     endem = re_End.match(lines[i])
 232                     if endem:
 233                         i += 1
 234                         break
 235                     i += 1
 236             i += 1
 237             continue
 238
 239         if format == 8:
 240             # We want to scan for ams-type includes and, if we find them,
 241             # add corresponding UseModule tags to the layout.
 242             match = re_AMSMaths.match(lines[i])
 243             if match:
 244                 addstring("theorems-ams", usemodules)
 245                 addstring("theorems-ams-extended", usemodules)
 246                 addstring("theorems-sec", usemodules)
 247                 lines.pop(i)
 248                 continue
 249             match = re_AMSMathsPlain.match(lines[i])
 250             if match:
 251                 addstring("theorems-starred", usemodules)
 252                 lines.pop(i)
 253                 continue
 254             match = re_AMSMathsSeq.match(lines[i])
 255             if match:
 256                 addstring("theorems-ams", usemodules)
 257                 addstring("theorems-ams-extended", usemodules)
 258                 lines.pop(i)
 259                 continue
 260             i += 1
 261             continue
 262
 263         # These just involved new features, not any changes to old ones
 264         if format >= 5 and format <= 7:
 265           i += 1
 266           continue
 267
 268         if format == 4:
 269             # Handle conversion to long CharStyle names
 270             match = re_CharStyle.match(lines[i])
 271             if match:
 272                 lines[i] = "InsetLayout CharStyle:%s" % (match.group(3))
 273                 i += 1
 274                 lines.insert(i, "\tLyXType charstyle")
 275                 i += 1
 276                 lines.insert(i, "")
 277                 lines[i] = "\tLabelString %s" % (match.group(3))
 278             i += 1
 279             continue
 280
 281         if format == 3:
 282             # convert 'providesamsmath x',  'providesmakeidx x',  'providesnatbib x',  'providesurl x' to
 283             #         'provides amsmath x', 'provides makeidx x', 'provides natbib x', 'provides url x'
 284             # x is either 0 or 1
 285             match = re_Provides.match(lines[i])
 286             if match:
 287                 lines[i] = "%sProvides %s%s%s" % (match.group(1), match.group(2).lower(),
 288                                                   match.group(3), match.group(4))
 289             i += 1
 290             continue
 291
 292         if format == 2:
 293             caption = []
 294
 295             # delete caption styles
 296             match = re_Style.match(lines[i])
 297             if match:
 298                 style = string.lower(match.group(4))
 299                 if style == "caption":
 300                     del lines[i]
 301                     while i < len(lines) and not re_End.match(lines[i]):
 302                         caption.append(lines[i])
 303                         del lines[i]
 304                     if i == len(lines):
 305                         error('Incomplete caption style.')
 306                     else:
 307                         del lines[i]
 308                         continue
 309
 310             # delete undefinition of caption styles
 311             match = re_NoStyle.match(lines[i])
 312             if match:
 313                 style = string.lower(match.group(4))
 314                 if style == "caption":
 315                     del lines[i]
 316                     continue
 317
 318             # replace the CopyStyle statement with the definition of the real
 319             # style. This may result in duplicate statements, but that is OK
 320             # since the second one will overwrite the first one.
 321             match = re_CopyStyle.match(lines[i])
 322             if match:
 323                 style = string.lower(match.group(4))
 324                 if style == "caption":
 325                     if len(caption) > 0:
 326                         lines[i:i+1] = caption
 327                     else:
 328                         # FIXME: This style comes from an include file, we
 329                         # should replace the real style and not this default.
 330                         lines[i:i+1] = ['       Margin                First_Dynamic',
 331                                         '       LatexType             Command',
 332                                         '       LatexName             caption',
 333                                         '       NeedProtect           1',
 334                                         '       LabelSep              xx',
 335                                         '       ParSkip               0.4',
 336                                         '       TopSep                0.5',
 337                                         '       Align                 Center',
 338                                         '       AlignPossible         Center',
 339                                         '       LabelType             Sensitive',
 340                                         '       LabelString           "Senseless!"',
 341                                         '       OptionalArgs          1',
 342                                         '       LabelFont',
 343                                         '         Series              Bold',
 344                                         '       EndFont']
 345
 346             i += 1
 347             continue
 348
 349         # Delete MaxCounter and remember the value of it
 350         match = re_MaxCounter.match(lines[i])
 351         if match:
 352             level = match.group(4)
 353             if string.lower(level) == "counter_chapter":
 354                 maxcounter = 0
 355             elif string.lower(level) == "counter_section":
 356                 maxcounter = 1
 357             elif string.lower(level) == "counter_subsection":
 358                 maxcounter = 2
 359             elif string.lower(level) == "counter_subsubsection":
 360                 maxcounter = 3
 361             elif string.lower(level) == "counter_paragraph":
 362                 maxcounter = 4
 363             elif string.lower(level) == "counter_subparagraph":
 364                 maxcounter = 5
 365             elif string.lower(level) == "counter_enumi":
 366                 maxcounter = 6
 367             elif string.lower(level) == "counter_enumii":
 368                 maxcounter = 7
 369             elif string.lower(level) == "counter_enumiii":
 370                 maxcounter = 8
 371             del lines[i]
 372             continue
 373
 374         # Replace line
 375         #
 376         # LabelType Counter_EnumI
 377         #
 378         # with two lines
 379         #
 380         # LabelType Counter
 381         # LabelCounter EnumI
 382         #
 383         match = re_LabelType.match(lines[i])
 384         if match:
 385             label = match.group(4)
 386             # Remember indenting space for later reuse in added lines
 387             space1 = match.group(1)
 388             # Remember the line for adding the LabelCounter later.
 389             # We can't do it here because it could shift latextype_line etc.
 390             labeltype_line = i
 391             if string.lower(label[:8]) == "counter_":
 392                 counter = string.lower(label[8:])
 393                 lines[i] = re_LabelType.sub(r'\1\2\3Counter', lines[i])
 394
 395         # Remember the LabelString line
 396         match = re_LabelString.match(lines[i])
 397         if match:
 398             labelstring = match.group(4)
 399             labelstring_line = i
 400
 401         # Remember the LabelStringAppendix line
 402         match = re_LabelStringAppendix.match(lines[i])
 403         if match:
 404             labelstringappendix = match.group(4)
 405             labelstringappendix_line = i
 406
 407         # Remember the LatexType line
 408         match = re_LatexType.match(lines[i])
 409         if match:
 410             latextype = string.lower(match.group(4))
 411             latextype_line = i
 412
 413         # Remember the TocLevel line
 414         match = re_TocLevel.match(lines[i])
 415         if match:
 416             toclevel = string.lower(match.group(4))
 417
 418         # Reset variables at the beginning of a style definition
 419         match = re_Style.match(lines[i])
 420         if match:
 421             style = string.lower(match.group(4))
 422             counter = ""
 423             toclevel = ""
 424             label = ""
 425             space1 = ""
 426             labelstring = ""
 427             labelstringappendix = ""
 428             labelstring_line = -1
 429             labelstringappendix_line = -1
 430             labeltype_line = -1
 431             latextype = ""
 432             latextype_line = -1
 433
 434         if re_End.match(lines[i]):
 435
 436             # Add a line "LatexType Bib_Environment" if LabelType is Bibliography
 437             # (or change the existing LatexType)
 438             if string.lower(label) == "bibliography":
 439                 if (latextype_line < 0):
 440                     lines.insert(i, "%sLatexType Bib_Environment" % space1)
 441                     i += 1
 442                 else:
 443                     lines[latextype_line] = re_LatexType.sub(r'\1\2\3Bib_Environment', lines[latextype_line])
 444
 445             # Change "LabelType Static" to "LabelType Itemize" for itemize environments
 446             if latextype == "item_environment" and string.lower(label) == "static":
 447                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Itemize', lines[labeltype_line])
 448
 449             # Change "LabelType Counter_EnumI" to "LabelType Enumerate" for enumerate environments
 450             if latextype == "item_environment" and string.lower(label) == "counter_enumi":
 451                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Enumerate', lines[labeltype_line])
 452                 # Don't add the LabelCounter line later
 453                 counter = ""
 454
 455             # Replace
 456             #
 457             # LabelString "Chapter"
 458             #
 459             # with
 460             #
 461             # LabelString "Chapter \arabic{chapter}"
 462             #
 463             # if this style has a counter. Ditto for LabelStringAppendix.
 464             # This emulates the hardcoded article style numbering of 1.3
 465             #
 466             if counter != "":
 467                 if counters.has_key(style):
 468                     if labelstring_line < 0:
 469                         lines.insert(i, '%sLabelString "%s"' % (space1, counters[style]))
 470                         i += 1
 471                     else:
 472                         new_labelstring = concatenate_label(labelstring, counters[style])
 473                         lines[labelstring_line] = re_LabelString.sub(
 474                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 475                                 lines[labelstring_line])
 476                 if appendixcounters.has_key(style):
 477                     if labelstringappendix_line < 0:
 478                         lines.insert(i, '%sLabelStringAppendix "%s"' % (space1, appendixcounters[style]))
 479                         i += 1
 480                     else:
 481                         new_labelstring = concatenate_label(labelstring, appendixcounters[style])
 482                         lines[labelstringappendix_line] = re_LabelStringAppendix.sub(
 483                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 484                                 lines[labelstringappendix_line])
 485
 486                 # Now we can safely add the LabelCounter line
 487                 lines.insert(labeltype_line + 1, "%sLabelCounter %s" % (space1, counter))
 488                 i += 1
 489
 490             # Add the TocLevel setting for sectioning styles
 491             if toclevel == "" and toclevels.has_key(style) and maxcounter <= toclevels[style]:
 492                 lines.insert(i, '%sTocLevel %d' % (space1, toclevels[style]))
 493                 i += 1
 494
 495         i += 1
 496
 497     if usemodules:
 498         i = formatline + 1
 499         for mod in usemodules:
 500             lines.insert(i, "UseModule " + mod)
 501             i += 1
 502
 503     return format + 1
 504
 505
 506 def main(argv):
 507
 508     # Open files
 509     if len(argv) == 1:
 510         input = sys.stdin
 511         output = sys.stdout
 512     elif len(argv) == 3:
 513         input = open(argv[1], 'rb')
 514         output = open(argv[2], 'wb')
 515     else:
 516         error(usage(argv[0]))
 517
 518     # Do the real work
 519     lines = read(input)
 520     format = 1
 521     while (format < currentFormat):
 522         format = convert(lines)
 523     write(output, lines)
 524
 525     # Close files
 526     if len(argv) == 3:
 527         input.close()
 528         output.close()
 529
 530     return 0
 531
 532
 533 if __name__ == "__main__":
 534     main(sys.argv)